Introducing /θ/ and /ð/ to languages with neither

A forum for all topics related to constructed languages
Post Reply
User avatar
LinguistCat
cuneiform
cuneiform
Posts: 133
Joined: Sat 06 May 2017, 06:48

Introducing /θ/ and /ð/ to languages with neither

Post by LinguistCat » Thu 12 Jul 2018, 23:28

I know that both /θ/ and /ð/ are rare cross-linguistically. I also know some common ways they derive from other sounds, as well as that they tend to be unstable often becoming /f/ or /s/ and /v/ or /z/ depending on voicing. But I have some specific questions regarding these sounds.

1. If "plain" oral stops have allophones that are fricatives in certain environments, would it be likely that prenasalized stops would have fricative allophones in the same environments, or would they be likely to be completely separate?
a. If both plain stops and prenasalized ones are likely to have fricative allophones in the same environments, and the fricative allophone of /t/ is [θ] or [ð] (depending on other factors or in free variation), and of /nd/ is [nð], would one be more stable than the other?
b. Would the presence of /s/ and /nz/ affect that?

2. In general, what environments are likely to fricativize stops? Or is it common enough of a sound change that I should ask what environments don't? Tthe language I'm deriving my conlang from is strictly CV, unless you count the prenasalized stops as nasal-stop combos.
a. Would palatalization make fricativization more likely?
b. Would velarization have any effects?
User avatar
Creyeditor
mongolian
mongolian
Posts: 4412
Joined: Tue 14 Aug 2012, 18:32

Re: Introducing /θ/ and /ð/ to languages with neither

Post by Creyeditor » Fri 13 Jul 2018, 00:38

LinguistCat wrote:
Thu 12 Jul 2018, 23:28
I know that both /θ/ and /ð/ are rare cross-linguistically. I also know some common ways they derive from other sounds, as well as that they tend to be unstable often becoming /f/ or /s/ and /v/ or /z/ depending on voicing. But I have some specific questions regarding these sounds.

1. If "plain" oral stops have allophones that are fricatives in certain environments, would it be likely that prenasalized stops would have fricative allophones in the same environments, or would they be likely to be completely separate?
a. If both plain stops and prenasalized ones are likely to have fricative allophones in the same environments, and the fricative allophone of /t/ is [θ] or [ð] (depending on other factors or in free variation), and of /nd/ is [nð], would one be more stable than the other?
b. Would the presence of /s/ and /nz/ affect that?

2. In general, what environments are likely to fricativize stops? Or is it common enough of a sound change that I should ask what environments don't? Tthe language I'm deriving my conlang from is strictly CV, unless you count the prenasalized stops as nasal-stop combos.
a. Would palatalization make fricativization more likely?
b. Would velarization have any effects?
1. Prenasalized stops are often excempt from becoming fricatives by allophonic rules, because languages try to avoid nasalized fricatives.
a. Even though your premise is kind of wrong, [θ] (and generally voiceless fricatives) are generally more stable than voiced ones.
b. The presence of /s/ makes [θ] more stable, because it could not merge with /s/ into one category without loosing constrast in minimal pairs
2. Intervocallically
a. A sequence of changes might turn stops into palatal affricates, which then can be easily fricativized.
b. I don't think it would, I am not sure though.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
spanick
greek
greek
Posts: 612
Joined: Thu 11 May 2017, 00:47
Location: California

Re: Introducing /θ/ and /ð/ to languages with neither

Post by spanick » Fri 13 Jul 2018, 00:57

I agree with creyeditor for the most part.

In response to (2), I think intervocalic fricativization is quite common, especially with voiced stops. It would only take a couple other sound changes to make that phonemic.

(2a) yes, palatization can lead to africates which can then be smoothed to fricatives. In particular, I'm thinking of Spanish here which had palatalized Latin <t> /t/ > /ts/ > /s̺/ > /θ/
Last edited by spanick on Fri 13 Jul 2018, 03:48, edited 1 time in total.
User avatar
LinguistCat
cuneiform
cuneiform
Posts: 133
Joined: Sat 06 May 2017, 06:48

Re: Introducing /θ/ and /ð/ to languages with neither

Post by LinguistCat » Fri 13 Jul 2018, 03:40

Thanks, this already helps a lot. [:D]
User avatar
Zekoslav
hieroglyphic
hieroglyphic
Posts: 62
Joined: Sat 07 Oct 2017, 15:54

Re: Introducing /θ/ and /ð/ to languages with neither

Post by Zekoslav » Fri 13 Jul 2018, 10:40

I agree with everything that's been said so far, and I've got some more interesting natlang examples:

1. Stops can also fricativize when only preceded by a vowel, (Hebrew, Aramaic, maybe Old French), even though intervocalic fricativization is more common.

2. They can also fricativize spontaneously, (Greek, eastern Iranian, Proto-Germanic, various Oceanic languages). When the language has a voicing or aspiration contrast, usually it's the voiced or aspirated stops that fricativize. Oceanic is an interesting counter-example, which proves that prenasalized stops are the most resistant to fricativization: Proto-Oceanic had a contrast between plain and prenasalized stops, with plain stops being voiceless. In most Oceanic languages, plain stops become fricatives, variously voiceless or voiced, while prenasalized stops remain stops even when they loose their prenasalization. There's a good summary of these changes in this paper.

3. When it comes to palatalization, I recall reading that Mandarin /f/ comes from earlier /pj/.
Languages:
:hrv: [:D], :bih: :srb: [;)], :eng: [:D], :fra: [:|], :lat: [:(], :deu: [:'(]

A linguistics enthusiast who would like to make a conlang, but can't decide what to call what.

- Tewanian languages
- Guide to Slavic accentuation
User avatar
Shemtov
runic
runic
Posts: 2692
Joined: Mon 29 Apr 2013, 03:06

Re: Introducing /θ/ and /ð/ to languages with neither

Post by Shemtov » Fri 13 Jul 2018, 22:45

spanick wrote:
Fri 13 Jul 2018, 00:57
I agree with creyeditor for the most part.

In response to (2), I think intervocalic fricativization is quite common, especially with voiced stops. It would only take a couple other sound changes to make that phonemic.

(2a) yes, palatization can lead to africates which can then be smoothed to fricatives. In particular, I'm thinking of Spanish here which had palatalized Latin <t> /t/ > /ts/ > /s̺/ > /θ/
PSlavic also had a "second" palatization where from PIE *ḱ>t͡ʂ>ʂ and *k _i>t͡ʃ>t͡s>s. You could do something similar-one palatizatios goes to t͡ʃ>t͡s>s, the other happens which the shift t͡ʃ>t͡s from C>t͡ʃ and then t͡s>s causes t͡ʃ>t͡s in the second palatization which could shift to tθ and then >θ.
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien
Porphyrogenitos
sinic
sinic
Posts: 206
Joined: Sat 21 Jul 2012, 07:01
Location: Buffalo, NY

Re: Introducing /θ/ and /ð/ to languages with neither

Post by Porphyrogenitos » Sun 15 Jul 2018, 07:22

Zekoslav wrote:
Fri 13 Jul 2018, 10:40
Oceanic is an interesting counter-example, which proves that prenasalized stops are the most resistant to fricativization: Proto-Oceanic had a contrast between plain and prenasalized stops, with plain stops being voiceless. In most Oceanic languages, plain stops become fricatives, variously voiceless or voiced, while prenasalized stops remain stops even when they loose their prenasalization. There's a good summary of these changes in this paper.
I love this so much, I've got to use it sometime.

Imagine if the language developed prenasalized stops through reanalysis of nasal codas, thus allowing for prenasalized stops across word boundaries, possibly resulting in a really neat fortition mutation. Like:

penku 'dog' / on penku 'the dog'
peᵑgu 'dog' / o‿ᵐbeᵑgu 'the dog'
feku 'dog' / o peku 'the dog'
User avatar
Creyeditor
mongolian
mongolian
Posts: 4412
Joined: Tue 14 Aug 2012, 18:32

Re: Introducing /θ/ and /ð/ to languages with neither

Post by Creyeditor » Sun 15 Jul 2018, 10:15

This is actually what has been reconstructed for Bantu, where we find alternations like l <-> nd because Proto-Bantu *d lenited to /l/ in a lot of Bantu languages.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
Shemtov
runic
runic
Posts: 2692
Joined: Mon 29 Apr 2013, 03:06

Re: Introducing /θ/ and /ð/ to languages with neither

Post by Shemtov » Sun 15 Jul 2018, 17:40

Zekoslav wrote:
Fri 13 Jul 2018, 10:40
. Oceanic is an interesting counter-example, which proves that prenasalized stops are the most resistant to fricativization: Proto-Oceanic had a contrast between plain and prenasalized stops, with plain stops being voiceless.
Talking of Oceanic, the Peripheral Papauan Tip Sudest (an Exonym) had the unconditional sound change *s>ð. Very little is known of the language's diachronics, so no one's sure how it happened.
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien
User avatar
Zekoslav
hieroglyphic
hieroglyphic
Posts: 62
Joined: Sat 07 Oct 2017, 15:54

Re: Introducing /θ/ and /ð/ to languages with neither

Post by Zekoslav » Mon 16 Jul 2018, 10:37

Porphyrogenitos wrote:
Sun 15 Jul 2018, 07:22
Zekoslav wrote:
Fri 13 Jul 2018, 10:40
Oceanic is an interesting counter-example, which proves that prenasalized stops are the most resistant to fricativization: Proto-Oceanic had a contrast between plain and prenasalized stops, with plain stops being voiceless. In most Oceanic languages, plain stops become fricatives, variously voiceless or voiced, while prenasalized stops remain stops even when they loose their prenasalization. There's a good summary of these changes in this paper.
I love this so much, I've got to use it sometime.

Imagine if the language developed prenasalized stops through reanalysis of nasal codas, thus allowing for prenasalized stops across word boundaries, possibly resulting in a really neat fortition mutation. Like:

penku 'dog' / on penku 'the dog'
peᵑgu 'dog' / o‿ᵐbeᵑgu 'the dog'
feku 'dog' / o peku 'the dog'
This is interesting and plausible (it essentially combines Insular Celtic and Oceanic developments), and in fact this is probably how Oceanic prenasalized stops developped in the first place, even though most instances of consonant mutation have been leveled.

If you look at the pronouns reconstructed in the paper, you can see that possessive suffixes regularly derive from subject/object clitics by prenasalization - this prenasalization seems to be the remnant of a genitive preposition, reconstructed as *mu according to Wikipedia.

I have to admit, as soon as I first saw that paper, I also wanted to include that development in my IE. conlang. [:D]
Languages:
:hrv: [:D], :bih: :srb: [;)], :eng: [:D], :fra: [:|], :lat: [:(], :deu: [:'(]

A linguistics enthusiast who would like to make a conlang, but can't decide what to call what.

- Tewanian languages
- Guide to Slavic accentuation
Post Reply